Computer-aided detection of lung fibrosis remains a difficult task due to the small vascular structures, scars, and fibrotic tissues that need to be identified and differentiated. A texture-based computer-aided diagnosis (CAD) system was implemented that automatically detects lung fibrosis. Our system uses high-resolution computed tomography (HRCT), advanced texture analysis, and support vector machine (SVM) committees to automatically and accurately detect lung fibrosis.

Our CAD system follows a five-stage pipeline that is comprised of: segmentation, texture analysis, training, classification, and display. On average, when using the suggested/default texture size and an optimized SVM committee system, a 90% accuracy has been observed with outr texture-based CAD system.

Publications: Jesus J. Caban, Jianhua Yao, N.A. Avila, J.R. Fontana, and V.C. Manganiello, "Texture-Based Computed-Aided Diagnosis System for Lung Fibrosis", Proceedings of the SPIE, Volume 6514, pp. 651439 (2007), SPIE Medical Imaging 2007.

Collaborator: Jianhua Yao (CC/NIH)

Computed-Aided Diagnosis of Breast Cancer using Dynamic MRI

Breat cancer is the most common invasive cancer among women, accounting for nearly one in three of cancer diagnoses in the United States. It is also the second leading cause of cancer deaths in the United States, with only lung cancer causing more deaths. In this project, 4D statistical histogram analysis was used to analyze the effects of Gadolinium contrast agent over time and automatically find specifc region of interest characteristic of breast cancer.

Poster: Kryt Chattrabhuti, Jesus J. Caban, and Jianhua Yao, "CAD of Breast Cancer on Dynamic MRI". NIH Summer Biomedical Research Review, NIH.

Collaborators: Jianhua Yao (CC/NIH) and Kryt Chattrabhuti (JHU)

A method to register 3D volumetric data with 2D transesophageal echocardiography (TEE) was developed. Our system uses TEE video analysis, landmark-based affine registration of TEE key frames, interpolation between key frames, and deformable registration to accomplish an accurate registration between TEE video with 3D volumetric data.

The fusion and insertion of 2D echocardiographic video within 3D volumetric data has proved to refine the overall understanding, interpretation, and orientation TEE procedures.

 Poster: Jesus J. Caban, Wendy K. Bernstein, Inna Shats, Ivan George, and Adrian Park, "Registration of 3D Volumes and Echocardiography Images for Training Purposes", Medicine Meets Virtual Reality (MMVR) 2007. Long Beach, California, 2007

 Collaborators: Wendy K. Bernstein (UMM) and Ivan George (UMMC)